A decade ago, we discovered that bright (high-intensity) light exposure could suppress nighttime melatonin production in humans. We were able to make this discovery partly, because of our newly-developed gas chromatographic - negative chemical ionization mass spectrometric assay for measurement of melatonin. This discovery reversed the thinking that humans were not affected by light in ways similar to those of other animals. Based on this finding, we used bright light to treat the first patient with winter depression. We and others have also begun to define the human phase response curve for light: bright light exposure in the morning causes phase advances (shifts to an earlier time), whereas bright light exposure in the evening causes circadian phase delays (shifts to a later time). Accordingly, we have hypothesized that there are two types of disorders of human circadian phase: phase-advanced disorders and phase-delayed disorders. In the former category are advanced sleep phase syndrome and jet lag after east-to-west travel. In the latter category are delayed sleep phase syndrome and jet lag after east-to-west travel. We have hypothesized that most winter depressives are in the latter category (the phase-shift hypothesis for winter depression and the antidepressants effect of light). A major thrust of our work over the last five years has been the testing of the phase-shift hypothesis. The current grant proposes to subject this hypothesis to two critical tests. One of these will be to evaluate the melatonin profile in winter depressives at all seasons of the year to determine if winter depressives are phase delayed when they are depressed in the winter compared to when they are euthymic in the summer. The second test is based on one of our most recent and most important discoveries. We have discovered that a physiological dose of appropriately-timed melatonin administration can cause robust and reproducible circadian phase shifts in sighted humans. We have, in fact, preliminarily described a melatonin phase response curve: melatonin administered in the morning causes phase delays; melatonin in the afternoon causes phase advances. We plan to subject the phase-shift hypothesis to a critical test by determining whether or not melatonin administered in the afternoon is more antidepressant than melatonin administered in the morning. The melatonin PRC needs to be more precisely defined before melatonin can be used to treat chronobiologic disorders. In addition to our work with winter depressives, we plan to continue developing melatonin as a treatment for free-running blind people and for sleep disorders in sighted individuals. Melatonin is potentially a sleeping aid without the side effects commonly associated with hypnotics and other psychotropic medications. Our work should be very important in the understanding of chronobiologic disorders, as well as in their treatment with appropriately-timed melatonin administration, alone or in conjunction with appropriately-timed bright light exposure.